df_education_rates <- df %>%
  filter(!is.na(avgBirthWomen), !is.na(popNoEduc), !is.na(popBasicEduc)) %>%
  mutate(totalEduPop = popNoEduc + popBasicEduc) %>%
  mutate(
    noEduRate = avgBirthWomen * (popNoEduc / totalEduPop),
    basicEduRate = avgBirthWomen * (popBasicEduc / totalEduPop)
  ) %>%
  select(Year, noEduRate, basicEduRate) %>%
  pivot_longer(
    cols = c(noEduRate, basicEduRate),
    names_to = "EducationLevel",
    values_to = "ApproxBirthRate"
  )

ui <- fluidPage(
  titlePanel("Birth Rates by Education Level Over Time (U.S.)"),
  
  sidebarLayout(
    sidebarPanel(
      sliderInput("yearRange", 
                  "Select Year Range:", 
                  min = min(df_education_rates$Year), 
                  max = max(df_education_rates$Year),
                  value = c(min(df_education_rates$Year), max(df_education_rates$Year)),
                  step = 1,
                  sep = ""),
      
      checkboxGroupInput("eduLevels", 
                         "Select Education Level(s):",
                         choices = list("No Education" = "noEduRate", "Basic Education" = "basicEduRate"),
                         selected = c("noEduRate", "basicEduRate"))
    ),
    
    mainPanel(
      plotlyOutput("birthRatePlot")
    )
  )
)

server <- function(input, output) {
  filteredData <- reactive({
    df_education_rates %>%
      filter(Year >= input$yearRange[1], Year <= input$yearRange[2]) %>%
      filter(EducationLevel %in% input$eduLevels)
  })
  
  output$birthRatePlot <- renderPlotly({
    p <- ggplot(filteredData(), aes(x = Year, y = ApproxBirthRate, color = EducationLevel)) +
      geom_line(size = 1.2) +
      geom_point(size = 2) +
      scale_color_manual(
        values = c(
          "noEduRate" = "#3498db",  # Blue
          "basicEduRate" = "#e74c3c"  # Red
        ),
        labels = c(
          "noEduRate" = "No Education",
          "basicEduRate" = "Basic Education"
        )
      ) +
      labs(
        x = "Year",
        y = "Estimated Births per Woman",
        color = "Education Level"
      ) +
      theme_minimal()
    
    ggplotly(p)
  })
}

#shinyApp(ui, server)

mergedWorld = df[18:87,] 
#keeps all columns but only keeps rows with GDP value

ui <- fluidPage(
  titlePanel("GDP vs Birthrates Comparison"),
  
  fluidRow(
    column(6, plotOutput("plot1", brush = brushOpts(id = "sharedBrush", resetOnNew = FALSE))),
    column(6, plotOutput("plot2", brush = brushOpts(id = "sharedBrush", resetOnNew = FALSE)))
  )
)

server <- function(input, output, session) {
  selectedYears <- reactiveVal(numeric(0))

  observeEvent(input$sharedBrush, {
    brushed <- brushedPoints(mergedWorld, input$sharedBrush, xvar = "Year", yvar = NULL)
    selectedYears(brushed$Year)
  })

  output$plot1 <- renderPlot({
    highlightYears <- selectedYears()
    mergedWorld %>%
      mutate(highlight = Year %in% highlightYears) %>%
      ggplot(aes(x = Year, y = avgBirthWomen)) +
      geom_point(aes(alpha = highlight, size = highlight)) +
      scale_alpha_manual(values = c("TRUE" = 1, "FALSE" = 0.5), guide = "none") +
      scale_size_manual(values = c("TRUE" = 3, "FALSE" = 1), guide = "none") +
      labs(title = "Birth Rates Over Time", y = "Average birth rate per woman")
  })

  output$plot2 <- renderPlot({
    highlightYears <- selectedYears()
    mergedWorld %>%
      mutate(highlight = Year %in% highlightYears) %>%
      ggplot(aes(x = Year, y = gpdCapita)) +
      geom_point(aes(alpha = highlight, size = highlight)) +
      scale_alpha_manual(values = c("TRUE" = 1, "FALSE" = 0.3), guide = "none") +
      scale_size_manual(values = c("TRUE" = 3, "FALSE" = 1), guide = "none") +
      labs(title = "GDP Per Capita Over Time", y = "GDP per Capita")
  })
}

#shinyApp(ui, server)

# read and clean data
# data = read_csv("data/mergedWorldDataset.csv", show_col_types = FALSE)
data = df |>
  filter(!is.na(avgChildbearingAge), !is.na(womenEmpowermentIndex), !is.na(marriageRate))|>
  select(Year, avgChildbearingAge, womenEmpowermentIndex, marriageRate)


# create layout with slider for year, and choices for y-axis metrics
ui = fluidPage(
  titlePanel("Women's Empowerment Data Over Time (U.S)"),
  sidebarLayout(
    sidebarPanel(
      sliderInput("yearRange", "Select Year Range", min(data$Year), max(data$Year), 
                  value = c(min(data$Year), max(data$Year)),step = 1, sep = ""),
      selectInput("yMetric", "Select Y-Axis Metric: ", 
                  choices = c("Average Childbearing Age" = "avgChildbearingAge","Marriage Rate" = "marriageRate"),
                  selected = "avgChildbearingAge")),
    mainPanel(plotlyOutput("barLinePlot"))
    
  )
)


# server that displays graph
server = function(input, output) {
  
  output$barLinePlot = renderPlotly({
    filtered_data = data |>
      filter(Year >=input$yearRange[1], Year <= input$yearRange[2]) |>
      mutate(
        selectedMetric = .data[[input$yMetric]], 
        emp = womenEmpowermentIndex
        )
    
    
    plot = ggplot(filtered_data, aes(x = Year)) +
      geom_col(aes(y = selectedMetric, fill = emp), width = 0.5) +
      scale_fill_gradient(low = "lightblue", high = "navy")+
      scale_x_continuous(breaks = seq(min(filtered_data$Year), max(filtered_data$Year), 5)) +
      labs(
        x = "Year",
        y = ifelse(input$yMetric == "avgChildbearingAge", "Average Childbearing Age", "Marriage Rate"),
        fill = "Empowerment Index",
        title = "Selected Metric vs. Women's Empowerment Over Time"
        ) +
      theme_minimal() +
      theme(axis.text.x = element_text(angle = 45, hjust = 1))
    
    ggplotly(plot)
  })
}


# run shiny app
#shinyApp(ui = ui, server = server)

# Focused variables
data_vars <- df %>% 
  select(
    Year,
    avgBirthWomen,
    democracyIndex,
    gpdCapita,
    womenEmpowermentIndex,
    population,
    marriageRate,
    popNoEduc,
    popBasicEduc
    ) %>%
  arrange(Year)